wdd-ls-map.png

The first step in developing a drug is identifying possible gene targets that could be related to a disease. If these genes can be effected in the right way, they may prove beneficial to curing or reducing symptoms of a disease. 

This product was originally developed as a technology solution (without design input) for biomedical researchers to search for gene targets through literature search, but users were having trouble understanding how to get the most out of the results. 

 

The challenge

Our challenge was to understand the user need and redesign the workflow of the tool to match their mental model and present results in a meaningful and easy to understand way.

 

Research

persona.png

We conducted a round of generative research to understand more about our main user. Our persona is responsible for validating the relationship between a disease and a gene. This step is crucial before moving to the next step of developing a molecule that can be tested. They can use many tools to assess the validity of certain targets, often running a few small scale experiments to generate some hypotheses and then turning to literature search to further validate the results. This is where Watson for Drug Discovery can help.

 
wdd-as-is.png

We performed several rounds of additional user research, evaluating the current experience of using the product to understand where the the biggest pain points arose. We found that overall, there was a lack of guidance for the user of which area of the product to use next and that it was too difficult to understand how the results ranked against each other in terms of the best possible potential targets.

 
wdd pain points.png
 
 

Solution

Our design solution gave the user a way to input their hypothesis in a way that was more familiar to them, then display results in one view without the user needing to perform multiple analyses in different parts of the product. This concept unlocked the power of the product while simplifying the overall experience an incredible amount.

 

Impact

This work is still in progress, but the feedback we’ve received from users of the product has been overwhelmingly positive.

"This [concept] is comprehensive and intuitive. We currently don’t have a way to do this in a unbiased fashion, so this is a very unbiased way for people from a variety of backgrounds to [validate targets using the literature].”

- User from Merck

"This is what I do and how I find targets or molecules. So I think this concept would work really well.”

- User from Stemonix